"This paper analyzes four German-distributed BioNTech/Pfizer BNT162b2 COVID-19 vaccine lots and reports that the injections not only transfect human HEK293 cells to produce spike protein for several days, but also contain measurable amounts of residual DNA far above international regulatory limits—including full plasmid elements such as the spike gene template, antibiotic resistance gene, origin of replication, and a duplicated SV40 promoter/enhancer sequence. The authors show that this DNA can enter human cells alongside the mRNA, and they state that spike protein is released mainly via exosomes, which could facilitate distribution to other tissues. Based on these findings, the paper argues that the presence and cellular uptake of plasmid DNA raise major safety concerns and calls for an immediate halt to RNA-based biologicals until these issues are resolved.
Provided below is a section-by-section overview of the paper "BioNTech RNA-Based COVID-19 Injections Contain Large Amounts of Residual DNA Including an SV40 Promoter/Enhancer Sequence"
https://www.semanticscholar.org/paper/BioNTech-RNA-Based-COVID-19-Injections-Contain-Of-K%C3%A4mmerer-Schulz/9f0e3c94329f877755a1e1ff7e960fa023a68d60
- Abstract
The authors state that Pfizer/BioNTech’s BNT162b2 vaccine successfully transfects HEK293 human cells, producing spike protein that is released mostly in exosome-type extracellular vesicles. They report finding large amounts of DNA in all lots tested (32.7–43.4 ng per dose), which they claim exceeds regulatory limits (10 ng/dose). PCR reportedly detected spike-coding DNA fragments, full plasmid genes, an SV40 promoter/enhancer, and the antibiotic resistance gene. They conclude this raises major safety concerns.
- Introduction
The paper outlines the origin of mRNA vaccines (Operation Warp Speed, Project Lightspeed). It describes the switch from process-1 (PCR-generated DNA templates) to process-2 (bacterial plasmid amplification) for large-scale mRNA production.
It summarizes prior reports that:
Spike protein was detectable in blood for weeks post-vaccination.
Spike protein found in tissues post-mortem and in placentas.
Vaccine mRNA fragments reported in blood cells.
modRNA uses pseudouridine and codon optimization to resist degradation.
Other studies suggested modRNA may undergo reverse transcription in vitro.
Previous analyses detected plasmid DNA—including SV40 enhancer sequences—in Pfizer lots.
The authors propose three questions:
Can they confirm high DNA levels in German BioNTech lots?
Can residual plasmid DNA be transfected into human cells?
Can the injections induce prolonged spike production?
Materials and Methods
3.1 Vaccine Lots
Four unopened BioNTech lots (three monovalent, one bivalent) were tested; one additional lot (GH9715) previously confirmed to contain plasmid DNA served as positive control.
3.2 Cell Line Experiments and ELISA
HEK293 cells were grown under standard conditions; vaccine vials were diluted to clinical dose levels. Spike levels in cell lysates and supernatants were measured on days 1, 3, 5, and 7.
3.3 Immunohistochemistry
Cells were transfected, fixed, embedded, stained with anti-spike antibodies, and visualized to confirm intracellular spike protein.
3.4 DNA and RNA Quantification
RNA quantified with Qubit RNA HS assay. DNA quantified using three assays: PicoGreen, Qubit dsDNA HS, and AccuBlue.
RNase A was used to remove RNA interference. Triton-X-100 used to open lipid nanoparticles (LNPs).
DNA/RNA ratios were compared to EMA allowable limits.
3.5 Polymerase Chain Reaction (PCR)
Primers targeting multiple areas of the production plasmid (SV40 enhancer, neomycin resistance gene, ORI replicon, spike gene) were designed.
PCR performed on DNA extracted from:
the vials
HEK293 cells after transfection
A previously sequenced plasmid-containing lot served as positive control.
3.6 Extracellular Vesicle (EV) Isolation
EVs were isolated using VN-96 peptide capture. Three fractions were analyzed: cell lysate, EVs, EV-free medium.
3.7 Sample Preparation for Proteomics
Cells, EVs, and supernatants were lysed, run through SDS-PAGE, cut, digested with trypsin, and prepared for mass spectrometry.
3.8 Mass Spectrometry-Based Proteomics
Samples run through LC-MS/MS using a TimsTOF-HT system. Data searched against human proteome, Moderna and Pfizer vector sequences, and SARS-CoV-2 spike.
- Results
4.1 Successful Transfection & Spike Protein Production
All four lots yielded strong spike staining in HEK293 cells.
Transfection efficiency: 74–90% depending on lot.
Cytopathic effects observed (vacuoles, cell detachment).
ELISA showed spike protein present from day 1, increasing through day 5, still elevated at day 7.
4.2 Spike Protein Release Into Supernatant
Spike was detected in culture medium (for monovalent lots), increasing over time.
Bivalent lot HD9869 had levels below detection due to assay only measuring Wuhan-type spike.
4.3 Spike Protein Primarily Released Through Extracellular Vesicles
Mass spectrometry showed spike protein distribution:
Lysate (most abundant)
EV fraction (detectable)
EV-free medium (minimal)
EV markers were enriched in EV fractions.
4.4 RNA Concentration Matches BioNTech Specification
Measured RNA levels ~26–28 µg per clinical dose, close to the declared 30 µg.
4.5 Large Amounts of DNA Found in All Lots
Without RNase:
DNA values varied widely depending on assay, ranging 1326–4225 ng per dose.
After RNase digestion (“true DNA”):
Residual DNA measured 32.7–43.4 ng per dose, in all lots.
The authors state this is 4–5× higher than EMA limit (10 ng/dose).
Opening LNPs with Triton-X-100 increased measurable DNA by 1.6–6.7×, implying DNA was inside LNPs.
4.6 Residual DNA Contains Full Plasmid Elements
PCR detected:
SV40 promoter/enhancer
Neomycin resistance gene
Origin of replication
Spike coding region
Multiple other regions across the entire plasmid map
These fragments were detected both:
in the vials, and
inside HEK293 cells after transfection, showing cellular uptake of plasmid DNA.
SV40 enhancer PCR showed two bands, consistent with two copies of the 72-bp element.
- Discussion
Key points stated by the authors:
Residual DNA levels in the tested lots exceed WHO and EMA limits.
Opening LNPs revealed more DNA, suggesting DNA was encapsulated.
Authors argue ethanol-precipitation methods in other studies may underestimate short DNA fragments.
They note existing regulatory limits were made for protein/antibody-based biologics, not LNP-packaged nucleic acids.
PCR detection of all plasmid elements—including SV40 promoter/enhancer—suggests process-related plasmid DNA was present and capable of entering cells.
Authors question BioNTech’s use of plasmids containing a mammalian-active SV40 nuclear localization enhancer.
They report that spike protein secretion occurred primarily via exosomes, which in vivo could transport spike to other tissues.
They reference preliminary unpublished work where another BNT162b2 lot reportedly transfected ovarian cancer cells with integration signals in chromosomes 9 and 12.
They report spike protein persists for days in vitro and increases through day 5.
They note toxicity of ionizable lipids and potential impacts on various cell types require further study.
- Limitations of the Study
Only a few lots were available.
Some vials were past expiration but stored at –80°C; authors argue integrity preserved.
Possibility of dsRNA or RNA:DNA hybrids present was not ruled out.
HEK293 cells are robust; results may differ in primary cells.
Full toxicity profile cannot be inferred.
- Conclusion
The authors conclude that:
All tested lots produce spike protein in human cells for several days.
Spike is released mainly via exosomes.
All lots contained residual plasmid DNA far above EMA limits.
All plasmid genes and SV40 promoter/enhancer elements were identified.
DNA entered and persisted inside cells.
They claim RNA biologicals pose four major dangers:
Autoimmunity from foreign protein expression
LNP toxicity
Genetic modification via plasmid DNA or reverse-transcribed RNA
Frameshifting from pseudouridine leading to unintended proteins
They call for an immediate halt to RNA-based injections."